Edit on BitBucket

Rimu.jl Package Guide

Random Integrators for many-body quantum systems

The grand aim is to develop a toolbox for many-body quantum systems that can be represented by a Hamiltonian in second quantisation language. Currently there are tools to find the ground state with FCIQMC or with a Lanczos algorithm (using KrylovKit for small Hilbert spaces). We will add tools to solve the time-dependent Schrödinger equation and Master equations for open system time evolution.

Contents

Installation

Install Rimu for running jobs only

Rimu can be installed with the package manager directly from the bitbucket repository. Either hit the ] key at the Julia REPL to get into Pkg mode and type

pkg> add https://joachimbrand@bitbucket.org/joachimbrand/rimu.jl#master

where master can be exchanged with the name of the desired git branch. Alternatively, use

julia> using Pkg; Pkg.add("https://joachimbrand@bitbucket.org/joachimbrand/rimu.jl#master")

Install Rimu for development

In order to be able to edit the source code, push changes, change and make new git branches, etc., clone the git repository with git clone to a convenient location, e.g. ~/mygitpackagefolder/. Then hit the ] key at the Julia REPL to get into Pkg mode and type

pkg> develop ~/mygitpackagefolder/rimu.jl

where the file path has to be adjusted to the location of the cloned git repository.

Usage

The package is now installed and can be imported with

julia> using Rimu

When planning to edit the code of the package it is advisable to use the Revise package by issuing

julia> using Revise

before using Rimu. This will track any changes made to the source code of Rimu and the changed package will be available after saving the source code (hopefully, and most of the time, without restarting the Julia REPL).

Rimu offers a number of tools for representing Hamiltonians (see Hamiltonians) and state vectors / wave functions (see DictVectors) as well as algorithms to find the ground state, e.g. fciqmc!.

Scripts

Rimu is written as a Julia package to be imported with using Rimu as described above. It supplies useful functions and types. Performing actual calculations and analysing the results is done with scripts. The folder scripts/ contains a collections of scripts that are either examples for use of the Rimu package or useful scripts for data analysis. In particular:

  • scripts/ros-BHM_M50_U6-W10k.jl is an example script that runs fciqmc on the 1D Bose-Hubbard model with MPI enabled. A data frame with results is written to the file fciqmcdata.feather.
  • scripts/read_file_and_plot.jl reads the feather file (from the working directory) and displays basic plots and blocking analysis of the shift.
  • plotting.jl is a collection of (currently very primitive) plotting function. On purpose these are not part of the Rimu package in order to avoid a dependency on a plotting package.

MPI

The Rimu package can run in parallel on different processes or node and distribute work by making use of MPI, or "message passing interface". For example, on my laptop, running

> julia scripts/ros-BHM_M50_U6-W10k.jl

will run on one processor with the main fciqmc!() computation (i.e. after package loading and compilation) completing in 18 seconds.

Running

> mpirun -np 4 julia scripts/ros-BHM_M50_U6-W10k.jl

on the same hardware makes use of 4 cores and the main part completes in 5.6 seconds, a speedup factor of 3.1. This seems reasonable, given that extra work needs to be done for communicating between different processes.

Initialising and finalising MPI communication has to be handled at the script level. Enabling MPI communication for use in fciqmc!() is done by wrapping the primary data structures as MPIData. A number of different strategies for data communication are implemented and most easily accessed with the functions:

See examples in the Scripts folder.